You don’t understand why wikis are leaving Fandom
It cuts deeper than simply ads. The reasons why editors are so dissatisfied reflect historical problems with the site.
Fandom has come under backlash in recent months, as some of its biggest wikis have changed hosting (the likes of the Minecraft and Zelda wiki).
To understand how we got here, we’ll go back almost two decades ago and follow the story of the Warcraft Wiki. See how it was lured to Fandom (formerly Wikia), how it eventually chose to leave it only to come back a few years later, and then once more go to a different host.
The wiki’s first years
The wiki started November 24, 2004, by Rustak as the go-to source on everything related to World of Warcraft.
For over 2 years, it ran on a budget by the founder, which meant that the server was increasingly harder for him to maintain. As such, he made the decision to go to Wikia:
As you’ve seen, I’ve had trouble keeping up with server issues, problems, and getting the upgrades and extensions that are needed to make WoWWiki better.
I’ve looked at a few options on how to solve this over the last month. When Jimmy Wales and Gil Penchina offered to let us move the wiki to Wikia, this seemed like a natural fit and I accepted. These guys have a record of treating communities well and running a great service, and I’m going to stay involved to help make sure that they deliver what our community needs.
Jimmy Wales was the founder of Wikipedia, and Wikia (later, Fandom) already hosted big names like Wookiepedia, Uncyclopedia and the Final Fantasy wiki with a great record. Best of all, it was free.
Of course it seemed like a natural fit!
But in october of 2010, Wikia made a breaking change to only have to work in one skin. Previously, wikis used the Monaco skin, and alternatively users could apply other skins (mainly Quartz and Monobook) for themselves. Now, wikis and users had only one skin: Oasis.
This was an enormous change, and as Wikia also enacted a more restrictive customization policy, the admins felt that it left little room for creating the wiki’s own identity.
Pcj, wiki admin, started a poll on whether to leave Wikia or not. He wrote:
WoWWiki (and myself personally) have nothing against progress or change, but sadly the “new” look is not conducive to the wiki experience — the focus is on getting traffic to other Wikia wikis.
After careful consideration, the community decided to leave.
But where?
Enter: Curse.
Curse — The Wikia competitor
Curse was more akin to a game managize than a wiki farm.
It was a collection of news, mods, tutorials, maps, game data, among other features. Unlike other sites, its business model was not reliant on wikis, but on building its communities and partnering with game developers.
One of their projects was to buy and host major community-ran sites, and among forums, databases and game extensions were wikis, which would later become Gamepedia.
It was exactly what WoWWiki needed.
We are being hosted by Curse. They are very excited to be working with us and we should see some nice improvements (tooltips auto-updated from Sigrie and the like) coming down the road. They are doing this ad-free (nice change from Wikia, huh?), at least for a few months. We did have a few other options — some of them very tempting — but the Curse option just clicked (and we had to move fast with Wikia shoving a [new] skin down our throats).
It’s critical to note here that Wikia refuses to close moved wikis.
As such, two competing wikis existed: WoWWiki (Wikia) and Wowpedia (Gamepedia), which is the main problem that plagues forked wikis not only then, but up to this day.
SEO is the issue. Newly created wikis take time to establish themselves and, as such, appear after the Fandom wiki in search engines. In the meantime, readers will see the Fandom one first and take it as the official one, which makes a negative feedback loop against the newly created wiki. Their only weapon is the support of the community.
Curse was able to provide this support by making publicity in their news articles, and Blizzard also helped ease the transition by linking to the new Warcraft wiki in their official sites.
This helped Wowpedia consolidate against WoWWiki in the long-term.
And for almost 10 years, Curse was home to the biggest and best gaming wikis in the world: Minecraft, Dota, Terraria, Don’t Starve, Zelda, Skyrim, World of Warcraft, you name it.
This was bad for Wikia.
Gamepedia was a direct competitor to Wikia, and as the latter made poor decisions, more wikis moved to the former. Pawel Dembowski, lead manager of Curse’s wiki team, said the following in an interview from 2014:
Wikia used to focus more on games, but today it’s more focused on finding new audiences and developing the social aspects of the wikis.
The actual content at Wikia is limited to less than half of the page because the rest is taken up by gamification. People come to Wikis to look up info or to learn more about the world of the game, not to get badges and stuff like that.
Fandom’s Discussions & Article Comments are the culmination of this.
Discussions is a forum-like feature that present a way for users to discuss everything from the wiki to general topics within the community. The problem with it is that it’s just social media. Why use Fandom’s forums when you can use Discord, Reddit, or Twitter?
Likewise, Article Comments are, well, comments you can add to pages. But the whole point of wikis is that they are factual and objective. Why, then, is it useful to say “Oh, this is my favorite character”? The only good usage of this feature is to highlight mistakes in the page, but that could already be done.
Other problems raised during this time:
- The introduction of videos that were in the page content and autoplayed. Bandwidth consumption and the annoyance of their presence in what was supposed to be a wiki resulted in strong backlash.
- The excessive amounts of ads that degraded from the reader’s experience, slowing down pages and many times appearing before the actual content of the pages.
- The inability to create an unique identity because they impose a very strict customization policy to create an uniform look across all wikis.
- The lack of communication with its editors. Every time a new experiment is ran, whether it be a new look, a new idea, or changes to existing features, they do it without consulting their wikis. And even if they do, they show no regard for their feedback.
For game wikis, Gamepedia was clearly the better hosting option.
Sparked by the backlash of auto-playing videos and Fandom’s lack of transparency, the sentiment to leave it turned more and more widespread.
And they were getting tired of it.
In december of 2018, Fandom officially announced the acquisition of Curse Media, which owned Gamepedia.
The cruel irony of being brought back to the very company you moved away from was a harsh one to swallow. At that moment, over 2,000 wikis began waiting to know what the future held for them.
Understanding the present situation
Fandom now had a development problem on their hands.
Fandom and Gamepedia’s underlying codebase were fundamentally different from each other. Fandom used a 7 year old version of MediaWiki (1.19) with heavy changes from the original software, while Gamepedia used version 1.31 with a very different look and feel to it.
They decided to create a new codebase for both platforms, called the Unified Community Platform (UCP), with the express intent of “bringing Fandom and Gamepedia wikis together on to the same version of MediaWiki, serving as the foundation upon which we can innovate”.
By early 2021, all wikis had been moved to it.
This had been a seamless transition for Gamepedia. What would not be seamless, however, was the follow-up.
The UCP was made in preparation for the new FandomDesktop and FandomMobile layouts (which, on a related note, took very little feedback), which meant not only a new (arguably worse) look, but also more ads, more experiments, more restrictions, and of course, dissent.
And Fandom still hasn’t worked out their transparency issues. The aforementioned problems have stayed plaguing Fandom, and now its new experiments started getting even bolder.
For example: This year, Fandom launched AI-generated Quick Answers. Why use AI when every single word of every page is already built by real people? It was incoherent, but worst of all, Fandom didn’t foretell wikis.
To quote the Hollow Knight wiki:
Not only were the questions themselves festering with improper grammar, but the answers themselves had details that were slid in from entirely unrelated video games and/or stories. The fact that we were not able to remove or edit them made it even worse. Fandom did not bother to test the AI that made these answers, and after this “feature” was added, we decided it was time to move from Fandom.
For Wowpedia, “the hosting situation with Fandom has reached a level that some of the admins feel is untenable”. “As a result, we are asking the community whether Wowpedia should [migrate].”
The results were a resounding “Yes”.
The new wiki provider was voted by the admins: wiki.gg.
To understand this new provider, we’ll see its parallels to Gamepedia.
Created by Gamepedia founder, Ben Robinson, and ex-Fandom president, Donovan Duncan, Freedom Games (owner of wiki.gg) is not a wiki farm, but a game publisher.
That’s their stroke of genius: they don’t have to monetise by using invasive ad networks like Google’s, or be at the mercy of goodwill like Wikipedia. They support themselves by promoting the games they publish.
And wiki.gg is simply a mean to that end.
Their interests are of keeping games and wikis under their host, which also means providing a good wiki farm, in constrast with Fandom, whose interest lie in keeping readers on their platform.
Hosting major game wikis like Terraria, Don’t Starve, Deep Rock Galactic, ARK and, more recently, the (now called) Warcraft Wiki, provide them with the public they need while giving a free and useful service to developers.
Closing thoughts
Change does not happen in a vaccum.
It wasn’t the amount of ads, or Quick Answers, or the skin change, or staff’s horrible transparency, or their customization policy, or their shift in focus to social aspects, or any other abitrary factor that made them move from Fandom, but it was the product of all of these over a long period of time.
Perhaps the biggest problem to Fandom was that Gamepedia wikis grew with a good semblance of independence. They chose the changes they wanted. Their voices were heard. They styled their wikis the way they wanted. The community ran the show.
And when wikis wouldn’t have it their way in Fandom, they left.
It’ll be interesting to see how Fandom will deal with their new competitor since Gamepedia didn’t sell themselves to Fandom, but their parent company, Twitch, did, because they were unprofitable. But now that Freedom Games is competely independent, it’s a different story.
Pcj, now Head of Wikis at Freedom Games, put it best:
Is there a guarantee that wiki.gg won’t be sold off? No, I don’t think you can predict that. [But] we don’t have any intention of doing so now or in the future. I know I won’t be going back to Fandom (again) in any case.
This is a still developing story, so we’ll have to keep an eye out on the steps Fandom (if any) will take to make sure they don’t lose their market share, as not only are wikis moving to wiki.gg, but there are major wikis going completely independent, notably Minecraft & Runecape wikis (Weird Gloop) and the Nintendo Independent Wikis Alliance (NIWA).
